blob: f7cedb5bc6f6cf2fd083566e62dc45ff8547df39 [file] [log] [blame]
// Copyright 2024 The Dawn & Tint Authors
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// 1. Redistributions of source code must retain the above copyright notice, this
// list of conditions and the following disclaimer.
//
// 2. Redistributions in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other materials provided with the distribution.
//
// 3. Neither the name of the copyright holder nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
// AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
// IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
// DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
// FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
// DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
// SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
// CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
// OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#include <utility>
#include "src/tint/lang/core/ir/transform/helper_test.h"
#include "src/tint/lang/core/type/struct.h"
#include "src/tint/lang/hlsl/writer/raise/shader_io.h"
namespace tint::hlsl::writer::raise {
namespace {
using namespace tint::core::fluent_types; // NOLINT
using namespace tint::core::number_suffixes; // NOLINT
using HlslWriterTransformTest = core::ir::transform::TransformTest;
TEST_F(HlslWriterTransformTest, ShaderIONoInputsOrOutputs) {
auto* ep = b.Function("foo", ty.void_(), core::ir::Function::PipelineStage::kCompute);
ep->SetWorkgroupSize(1, 1, 1);
b.Append(ep->Block(), [&] { b.Return(ep); });
auto* src = R"(
%foo = @compute @workgroup_size(1, 1, 1) func():void {
$B1: {
ret
}
}
)";
EXPECT_EQ(src, str());
auto* expect = src;
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_NonStruct) {
auto* front_facing = b.FunctionParam("front_facing", ty.bool_());
front_facing->SetBuiltin(core::BuiltinValue::kFrontFacing);
auto* position = b.FunctionParam("position", ty.vec4<f32>());
position->SetBuiltin(core::BuiltinValue::kPosition);
position->SetInvariant(true);
auto* color1 = b.FunctionParam("color1", ty.f32());
color1->SetLocation(0);
auto* color2 = b.FunctionParam("color2", ty.f32());
color2->SetLocation(1);
color2->SetInterpolation(core::Interpolation{core::InterpolationType::kLinear,
core::InterpolationSampling::kSample});
auto* ep = b.Function("foo", ty.void_(), core::ir::Function::PipelineStage::kFragment);
ep->SetParams({front_facing, position, color1, color2});
b.Append(ep->Block(), [&] {
auto* ifelse = b.If(front_facing);
b.Append(ifelse->True(), [&] {
b.Multiply(ty.vec4<f32>(), position, b.Add(ty.f32(), color1, color2));
b.ExitIf(ifelse);
});
b.Return(ep);
});
auto* src = R"(
%foo = @fragment func(%front_facing:bool [@front_facing], %position:vec4<f32> [@invariant, @position], %color1:f32 [@location(0)], %color2:f32 [@location(1), @interpolate(linear, sample)]):void {
$B1: {
if %front_facing [t: $B2] { # if_1
$B2: { # true
%6:f32 = add %color1, %color2
%7:vec4<f32> = mul %position, %6
exit_if # if_1
}
}
ret
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
foo_inputs = struct @align(16) {
color1:f32 @offset(0), @location(0)
color2:f32 @offset(4), @location(1), @interpolate(linear, sample)
position:vec4<f32> @offset(16), @invariant, @builtin(position)
front_facing:bool @offset(32), @builtin(front_facing)
}
%foo_inner = func(%front_facing:bool, %position:vec4<f32>, %color1:f32, %color2:f32):void {
$B1: {
if %front_facing [t: $B2] { # if_1
$B2: { # true
%6:f32 = add %color1, %color2
%7:vec4<f32> = mul %position, %6
exit_if # if_1
}
}
ret
}
}
%foo = @fragment func(%inputs:foo_inputs):void {
$B3: {
%10:bool = access %inputs, 3u
%11:vec4<f32> = access %inputs, 2u
%12:f32 = access %11, 3u
%13:f32 = div 1.0f, %12
%14:vec3<f32> = swizzle %11, xyz
%15:vec4<f32> = construct %14, %13
%16:f32 = access %inputs, 0u
%17:f32 = access %inputs, 1u
%18:void = call %foo_inner, %10, %15, %16, %17
ret
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_Struct) {
auto* str_ty = ty.Struct(mod.symbols.New("Inputs"),
{
{
mod.symbols.New("front_facing"),
ty.bool_(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kFrontFacing,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
{
mod.symbols.New("position"),
ty.vec4<f32>(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kPosition,
/* interpolation */ std::nullopt,
/* invariant */ true,
},
},
{
mod.symbols.New("color1"),
ty.f32(),
core::IOAttributes{
/* location */ 0u,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
{
mod.symbols.New("color2"),
ty.f32(),
core::IOAttributes{
/* location */ 1u,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */
core::Interpolation{
core::InterpolationType::kLinear,
core::InterpolationSampling::kSample,
},
/* invariant */ false,
},
},
});
auto* str_param = b.FunctionParam("inputs", str_ty);
auto* ep = b.Function("foo", ty.void_(), core::ir::Function::PipelineStage::kFragment);
ep->SetParams({str_param});
b.Append(ep->Block(), [&] {
auto* ifelse = b.If(b.Access(ty.bool_(), str_param, 0_i));
b.Append(ifelse->True(), [&] {
auto* position = b.Access(ty.vec4<f32>(), str_param, 1_i);
auto* color1 = b.Access(ty.f32(), str_param, 2_i);
auto* color2 = b.Access(ty.f32(), str_param, 3_i);
b.Multiply(ty.vec4<f32>(), position, b.Add(ty.f32(), color1, color2));
b.ExitIf(ifelse);
});
b.Return(ep);
});
auto* src = R"(
Inputs = struct @align(16) {
front_facing:bool @offset(0), @builtin(front_facing)
position:vec4<f32> @offset(16), @invariant, @builtin(position)
color1:f32 @offset(32), @location(0)
color2:f32 @offset(36), @location(1), @interpolate(linear, sample)
}
%foo = @fragment func(%inputs:Inputs):void {
$B1: {
%3:bool = access %inputs, 0i
if %3 [t: $B2] { # if_1
$B2: { # true
%4:vec4<f32> = access %inputs, 1i
%5:f32 = access %inputs, 2i
%6:f32 = access %inputs, 3i
%7:f32 = add %5, %6
%8:vec4<f32> = mul %4, %7
exit_if # if_1
}
}
ret
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Inputs = struct @align(16) {
front_facing:bool @offset(0)
position:vec4<f32> @offset(16)
color1:f32 @offset(32)
color2:f32 @offset(36)
}
foo_inputs = struct @align(16) {
Inputs_color1:f32 @offset(0), @location(0)
Inputs_color2:f32 @offset(4), @location(1), @interpolate(linear, sample)
Inputs_position:vec4<f32> @offset(16), @invariant, @builtin(position)
Inputs_front_facing:bool @offset(32), @builtin(front_facing)
}
%foo_inner = func(%inputs:Inputs):void {
$B1: {
%3:bool = access %inputs, 0i
if %3 [t: $B2] { # if_1
$B2: { # true
%4:vec4<f32> = access %inputs, 1i
%5:f32 = access %inputs, 2i
%6:f32 = access %inputs, 3i
%7:f32 = add %5, %6
%8:vec4<f32> = mul %4, %7
exit_if # if_1
}
}
ret
}
}
%foo = @fragment func(%inputs_1:foo_inputs):void { # %inputs_1: 'inputs'
$B3: {
%11:bool = access %inputs_1, 3u
%12:vec4<f32> = access %inputs_1, 2u
%13:f32 = access %12, 3u
%14:f32 = div 1.0f, %13
%15:vec3<f32> = swizzle %12, xyz
%16:vec4<f32> = construct %15, %14
%17:f32 = access %inputs_1, 0u
%18:f32 = access %inputs_1, 1u
%19:Inputs = construct %11, %16, %17, %18
%20:void = call %foo_inner, %19
ret
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_Mixed) {
auto* str_ty = ty.Struct(mod.symbols.New("Inputs"),
{
{
mod.symbols.New("position"),
ty.vec4<f32>(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kPosition,
/* interpolation */ std::nullopt,
/* invariant */ true,
},
},
{
mod.symbols.New("color1"),
ty.f32(),
core::IOAttributes{
/* location */ 0u,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
});
auto* front_facing = b.FunctionParam("front_facing", ty.bool_());
front_facing->SetBuiltin(core::BuiltinValue::kFrontFacing);
auto* str_param = b.FunctionParam("inputs", str_ty);
auto* color2 = b.FunctionParam("color2", ty.f32());
color2->SetLocation(1);
color2->SetInterpolation(core::Interpolation{core::InterpolationType::kLinear,
core::InterpolationSampling::kSample});
auto* ep = b.Function("foo", ty.void_(), core::ir::Function::PipelineStage::kFragment);
ep->SetParams({front_facing, str_param, color2});
b.Append(ep->Block(), [&] {
auto* ifelse = b.If(front_facing);
b.Append(ifelse->True(), [&] {
auto* position = b.Access(ty.vec4<f32>(), str_param, 0_i);
auto* color1 = b.Access(ty.f32(), str_param, 1_i);
b.Multiply(ty.vec4<f32>(), position, b.Add(ty.f32(), color1, color2));
b.ExitIf(ifelse);
});
b.Return(ep);
});
auto* src = R"(
Inputs = struct @align(16) {
position:vec4<f32> @offset(0), @invariant, @builtin(position)
color1:f32 @offset(16), @location(0)
}
%foo = @fragment func(%front_facing:bool [@front_facing], %inputs:Inputs, %color2:f32 [@location(1), @interpolate(linear, sample)]):void {
$B1: {
if %front_facing [t: $B2] { # if_1
$B2: { # true
%5:vec4<f32> = access %inputs, 0i
%6:f32 = access %inputs, 1i
%7:f32 = add %6, %color2
%8:vec4<f32> = mul %5, %7
exit_if # if_1
}
}
ret
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Inputs = struct @align(16) {
position:vec4<f32> @offset(0)
color1:f32 @offset(16)
}
foo_inputs = struct @align(16) {
Inputs_color1:f32 @offset(0), @location(0)
color2:f32 @offset(4), @location(1), @interpolate(linear, sample)
Inputs_position:vec4<f32> @offset(16), @invariant, @builtin(position)
front_facing:bool @offset(32), @builtin(front_facing)
}
%foo_inner = func(%front_facing:bool, %inputs:Inputs, %color2:f32):void {
$B1: {
if %front_facing [t: $B2] { # if_1
$B2: { # true
%5:vec4<f32> = access %inputs, 0i
%6:f32 = access %inputs, 1i
%7:f32 = add %6, %color2
%8:vec4<f32> = mul %5, %7
exit_if # if_1
}
}
ret
}
}
%foo = @fragment func(%inputs_1:foo_inputs):void { # %inputs_1: 'inputs'
$B3: {
%11:bool = access %inputs_1, 3u
%12:vec4<f32> = access %inputs_1, 2u
%13:f32 = access %12, 3u
%14:f32 = div 1.0f, %13
%15:vec3<f32> = swizzle %12, xyz
%16:vec4<f32> = construct %15, %14
%17:f32 = access %inputs_1, 0u
%18:Inputs = construct %16, %17
%19:f32 = access %inputs_1, 1u
%20:void = call %foo_inner, %11, %18, %19
ret
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOReturnValue_NonStructBuiltin) {
auto* ep = b.Function("foo", ty.vec4<f32>(), core::ir::Function::PipelineStage::kVertex);
ep->SetReturnBuiltin(core::BuiltinValue::kPosition);
ep->SetReturnInvariant(true);
b.Append(ep->Block(), [&] { b.Return(ep, b.Construct(ty.vec4<f32>(), 0.5_f)); });
auto* src = R"(
%foo = @vertex func():vec4<f32> [@invariant, @position] {
$B1: {
%2:vec4<f32> = construct 0.5f
ret %2
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
foo_outputs = struct @align(16) {
tint_symbol:vec4<f32> @offset(0), @invariant, @builtin(position)
}
%foo_inner = func():vec4<f32> {
$B1: {
%2:vec4<f32> = construct 0.5f
ret %2
}
}
%foo = @vertex func():foo_outputs {
$B2: {
%4:vec4<f32> = call %foo_inner
%5:foo_outputs = construct %4
ret %5
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOReturnValue_NonStructLocation) {
auto* ep = b.Function("foo", ty.vec4<f32>(), core::ir::Function::PipelineStage::kFragment);
ep->SetReturnLocation(1u);
b.Append(ep->Block(), [&] { b.Return(ep, b.Construct(ty.vec4<f32>(), 0.5_f)); });
auto* src = R"(
%foo = @fragment func():vec4<f32> [@location(1)] {
$B1: {
%2:vec4<f32> = construct 0.5f
ret %2
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
foo_outputs = struct @align(16) {
tint_symbol:vec4<f32> @offset(0), @location(1)
}
%foo_inner = func():vec4<f32> {
$B1: {
%2:vec4<f32> = construct 0.5f
ret %2
}
}
%foo = @fragment func():foo_outputs {
$B2: {
%4:vec4<f32> = call %foo_inner
%5:foo_outputs = construct %4
ret %5
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOReturnValue_Struct) {
auto* str_ty = ty.Struct(mod.symbols.New("Outputs"),
{
{
mod.symbols.New("position"),
ty.vec4<f32>(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kPosition,
/* interpolation */ std::nullopt,
/* invariant */ true,
},
},
{
mod.symbols.New("color1"),
ty.f32(),
core::IOAttributes{
/* location */ 0u,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
{
mod.symbols.New("color2"),
ty.f32(),
core::IOAttributes{
/* location */ 1u,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */
core::Interpolation{
core::InterpolationType::kLinear,
core::InterpolationSampling::kSample,
},
/* invariant */ false,
},
},
});
auto* ep = b.Function("foo", str_ty, core::ir::Function::PipelineStage::kVertex);
b.Append(ep->Block(), [&] {
b.Return(ep, b.Construct(str_ty, b.Construct(ty.vec4<f32>(), 0_f), 0.25_f, 0.75_f));
});
auto* src = R"(
Outputs = struct @align(16) {
position:vec4<f32> @offset(0), @invariant, @builtin(position)
color1:f32 @offset(16), @location(0)
color2:f32 @offset(20), @location(1), @interpolate(linear, sample)
}
%foo = @vertex func():Outputs {
$B1: {
%2:vec4<f32> = construct 0.0f
%3:Outputs = construct %2, 0.25f, 0.75f
ret %3
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Outputs = struct @align(16) {
position:vec4<f32> @offset(0)
color1:f32 @offset(16)
color2:f32 @offset(20)
}
foo_outputs = struct @align(16) {
Outputs_color1:f32 @offset(0), @location(0)
Outputs_color2:f32 @offset(4), @location(1), @interpolate(linear, sample)
Outputs_position:vec4<f32> @offset(16), @invariant, @builtin(position)
}
%foo_inner = func():Outputs {
$B1: {
%2:vec4<f32> = construct 0.0f
%3:Outputs = construct %2, 0.25f, 0.75f
ret %3
}
}
%foo = @vertex func():foo_outputs {
$B2: {
%5:Outputs = call %foo_inner
%6:vec4<f32> = access %5, 0u
%7:f32 = access %5, 1u
%8:f32 = access %5, 2u
%9:foo_outputs = construct %7, %8, %6
ret %9
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOReturnValue_DualSourceBlending) {
auto* str_ty =
ty.Struct(mod.symbols.New("Output"), {
{
mod.symbols.New("color1"),
ty.f32(),
core::IOAttributes{
/* location */ 0u,
/* blend_src */ 0u,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
{
mod.symbols.New("color2"),
ty.f32(),
core::IOAttributes{
/* location */ 0u,
/* blend_src */ 1u,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
});
auto* ep = b.Function("foo", str_ty, core::ir::Function::PipelineStage::kFragment);
b.Append(ep->Block(), [&] { b.Return(ep, b.Construct(str_ty, 0.25_f, 0.75_f)); });
auto* src = R"(
Output = struct @align(4) {
color1:f32 @offset(0), @location(0)
color2:f32 @offset(4), @location(0)
}
%foo = @fragment func():Output {
$B1: {
%2:Output = construct 0.25f, 0.75f
ret %2
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Output = struct @align(4) {
color1:f32 @offset(0)
color2:f32 @offset(4)
}
foo_outputs = struct @align(4) {
Output_color1:f32 @offset(0), @location(0)
Output_color2:f32 @offset(4), @location(0)
}
%foo_inner = func():Output {
$B1: {
%2:Output = construct 0.25f, 0.75f
ret %2
}
}
%foo = @fragment func():foo_outputs {
$B2: {
%4:Output = call %foo_inner
%5:f32 = access %4, 0u
%6:f32 = access %4, 1u
%7:foo_outputs = construct %5, %6
ret %7
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOStruct_SharedByVertexAndFragment) {
auto* str_ty = ty.Struct(mod.symbols.New("Interface"),
{
{
mod.symbols.New("position"),
ty.vec4<f32>(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kPosition,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
{
mod.symbols.New("color"),
ty.vec3<f32>(),
core::IOAttributes{
/* location */ 0u,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
});
// Vertex shader.
{
auto* ep = b.Function("vert", str_ty, core::ir::Function::PipelineStage::kVertex);
b.Append(ep->Block(), [&] {
auto* position = b.Construct(ty.vec4<f32>(), 0_f);
auto* color = b.Construct(ty.vec3<f32>(), 1_f);
b.Return(ep, b.Construct(str_ty, position, color));
});
}
// Fragment shader.
{
auto* inputs = b.FunctionParam("inputs", str_ty);
auto* ep = b.Function("frag", ty.vec4<f32>(), core::ir::Function::PipelineStage::kFragment);
ep->SetParams({inputs});
ep->SetReturnLocation(0u);
b.Append(ep->Block(), [&] {
auto* position = b.Access(ty.vec4<f32>(), inputs, 0_u);
auto* color = b.Access(ty.vec3<f32>(), inputs, 1_u);
b.Return(ep, b.Add(ty.vec4<f32>(), position, b.Construct(ty.vec4<f32>(), color, 1_f)));
});
}
auto* src = R"(
Interface = struct @align(16) {
position:vec4<f32> @offset(0), @builtin(position)
color:vec3<f32> @offset(16), @location(0)
}
%vert = @vertex func():Interface {
$B1: {
%2:vec4<f32> = construct 0.0f
%3:vec3<f32> = construct 1.0f
%4:Interface = construct %2, %3
ret %4
}
}
%frag = @fragment func(%inputs:Interface):vec4<f32> [@location(0)] {
$B2: {
%7:vec4<f32> = access %inputs, 0u
%8:vec3<f32> = access %inputs, 1u
%9:vec4<f32> = construct %8, 1.0f
%10:vec4<f32> = add %7, %9
ret %10
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Interface = struct @align(16) {
position:vec4<f32> @offset(0)
color:vec3<f32> @offset(16)
}
vert_outputs = struct @align(16) {
Interface_color:vec3<f32> @offset(0), @location(0)
Interface_position:vec4<f32> @offset(16), @builtin(position)
}
frag_inputs = struct @align(16) {
Interface_color:vec3<f32> @offset(0), @location(0)
Interface_position:vec4<f32> @offset(16), @builtin(position)
}
frag_outputs = struct @align(16) {
tint_symbol:vec4<f32> @offset(0), @location(0)
}
%vert_inner = func():Interface {
$B1: {
%2:vec4<f32> = construct 0.0f
%3:vec3<f32> = construct 1.0f
%4:Interface = construct %2, %3
ret %4
}
}
%frag_inner = func(%inputs:Interface):vec4<f32> {
$B2: {
%7:vec4<f32> = access %inputs, 0u
%8:vec3<f32> = access %inputs, 1u
%9:vec4<f32> = construct %8, 1.0f
%10:vec4<f32> = add %7, %9
ret %10
}
}
%vert = @vertex func():vert_outputs {
$B3: {
%12:Interface = call %vert_inner
%13:vec4<f32> = access %12, 0u
%14:vec3<f32> = access %12, 1u
%15:vert_outputs = construct %14, %13
ret %15
}
}
%frag = @fragment func(%inputs_1:frag_inputs):frag_outputs { # %inputs_1: 'inputs'
$B4: {
%18:vec4<f32> = access %inputs_1, 1u
%19:f32 = access %18, 3u
%20:f32 = div 1.0f, %19
%21:vec3<f32> = swizzle %18, xyz
%22:vec4<f32> = construct %21, %20
%23:vec3<f32> = access %inputs_1, 0u
%24:Interface = construct %22, %23
%25:vec4<f32> = call %frag_inner, %24
%26:frag_outputs = construct %25
ret %26
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOStruct_SharedWithBuffer) {
auto* str_ty = ty.Struct(mod.symbols.New("Outputs"),
{
{
mod.symbols.New("position"),
ty.vec4<f32>(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kPosition,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
{
mod.symbols.New("color"),
ty.vec3<f32>(),
core::IOAttributes{
/* location */ 0u,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
});
auto* var = b.Var(ty.ptr(storage, str_ty, read));
var->SetBindingPoint(0, 0);
auto* buffer = mod.root_block->Append(var);
auto* ep = b.Function("vert", str_ty, core::ir::Function::PipelineStage::kVertex);
b.Append(ep->Block(), [&] { b.Return(ep, b.Load(buffer)); });
auto* src = R"(
Outputs = struct @align(16) {
position:vec4<f32> @offset(0), @builtin(position)
color:vec3<f32> @offset(16), @location(0)
}
$B1: { # root
%1:ptr<storage, Outputs, read> = var @binding_point(0, 0)
}
%vert = @vertex func():Outputs {
$B2: {
%3:Outputs = load %1
ret %3
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Outputs = struct @align(16) {
position:vec4<f32> @offset(0)
color:vec3<f32> @offset(16)
}
vert_outputs = struct @align(16) {
Outputs_color:vec3<f32> @offset(0), @location(0)
Outputs_position:vec4<f32> @offset(16), @builtin(position)
}
$B1: { # root
%1:ptr<storage, Outputs, read> = var @binding_point(0, 0)
}
%vert_inner = func():Outputs {
$B2: {
%3:Outputs = load %1
ret %3
}
}
%vert = @vertex func():vert_outputs {
$B3: {
%5:Outputs = call %vert_inner
%6:vec4<f32> = access %5, 0u
%7:vec3<f32> = access %5, 1u
%8:vert_outputs = construct %7, %6
ret %8
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
// Test that IO attributes are stripped from structures that are not used for the shader interface.
TEST_F(HlslWriterTransformTest, ShaderIOStructWithAttributes_NotUsedForInterface) {
auto* vec4f = ty.vec4<f32>();
auto* str_ty = ty.Struct(mod.symbols.New("Outputs"),
{
{
mod.symbols.New("position"),
vec4f,
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kPosition,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
{
mod.symbols.New("color"),
vec4f,
core::IOAttributes{
/* location */ 0u,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ std::nullopt,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
});
auto* var = b.Var(ty.ptr(storage, str_ty, read));
var->SetBindingPoint(0, 0);
auto* buffer = mod.root_block->Append(var);
auto* ep = b.Function("frag", ty.void_(), core::ir::Function::PipelineStage::kFragment);
b.Append(ep->Block(), [&] {
b.Store(buffer, b.Construct(str_ty));
b.Return(ep);
});
auto* src = R"(
Outputs = struct @align(16) {
position:vec4<f32> @offset(0), @builtin(position)
color:vec4<f32> @offset(16), @location(0)
}
$B1: { # root
%1:ptr<storage, Outputs, read> = var @binding_point(0, 0)
}
%frag = @fragment func():void {
$B2: {
%3:Outputs = construct
store %1, %3
ret
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Outputs = struct @align(16) {
position:vec4<f32> @offset(0)
color:vec4<f32> @offset(16)
}
$B1: { # root
%1:ptr<storage, Outputs, read> = var @binding_point(0, 0)
}
%frag = @fragment func():void {
$B2: {
%3:Outputs = construct
store %1, %3
ret
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOCompute) {
auto* invoc = b.FunctionParam("invoc_id", ty.vec3<u32>());
invoc->SetBuiltin(core::BuiltinValue::kLocalInvocationId);
auto* ep = b.Function("cmp", ty.void_(), core::ir::Function::PipelineStage::kCompute);
ep->SetParams({invoc});
ep->SetWorkgroupSize(1, 1, 1);
b.Append(ep->Block(), [&] {
b.Let("a", invoc);
b.Return(ep);
});
auto* src = R"(
%cmp = @compute @workgroup_size(1, 1, 1) func(%invoc_id:vec3<u32> [@local_invocation_id]):void {
$B1: {
%a:vec3<u32> = let %invoc_id
ret
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
cmp_inputs = struct @align(16) {
invoc_id:vec3<u32> @offset(0), @builtin(local_invocation_id)
}
%cmp_inner = func(%invoc_id:vec3<u32>):void {
$B1: {
%a:vec3<u32> = let %invoc_id
ret
}
}
%cmp = @compute @workgroup_size(1, 1, 1) func(%inputs:cmp_inputs):void {
$B2: {
%6:vec3<u32> = access %inputs, 0u
%7:void = call %cmp_inner, %6
ret
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_Subgroup_NonStruct) {
auto* subgroup_invocation_id = b.FunctionParam("id", ty.u32());
subgroup_invocation_id->SetBuiltin(core::BuiltinValue::kSubgroupInvocationId);
auto* subgroup_size = b.FunctionParam("size", ty.u32());
subgroup_size->SetBuiltin(core::BuiltinValue::kSubgroupSize);
auto* ep = b.Function("foo", ty.u32(), core::ir::Function::PipelineStage::kFragment);
ep->SetParams({subgroup_invocation_id, subgroup_size});
b.Append(ep->Block(), [&] {
auto* r = b.Multiply(ty.u32(), subgroup_invocation_id, subgroup_size);
b.Return(ep, r);
});
auto* src = R"(
%foo = @fragment func(%id:u32 [@subgroup_invocation_id], %size:u32 [@subgroup_size]):u32 {
$B1: {
%4:u32 = mul %id, %size
ret %4
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
foo_outputs = struct @align(4) {
tint_symbol:u32 @offset(0)
}
%foo_inner = func(%id:u32, %size:u32):u32 {
$B1: {
%4:u32 = mul %id, %size
ret %4
}
}
%foo = @fragment func():foo_outputs {
$B2: {
%6:u32 = hlsl.WaveGetLaneIndex
%7:u32 = hlsl.WaveGetLaneCount
%8:u32 = call %foo_inner, %6, %7
%9:foo_outputs = construct %8
ret %9
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_Subgroup_Struct) {
auto* str_ty = ty.Struct(mod.symbols.New("Inputs"),
{
{
mod.symbols.New("id"),
ty.u32(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kSubgroupInvocationId,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
{
mod.symbols.New("size"),
ty.u32(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kSubgroupSize,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
});
auto* str_param = b.FunctionParam("inputs", str_ty);
auto* ep = b.Function("foo", ty.u32(), core::ir::Function::PipelineStage::kFragment);
ep->SetParams({str_param});
b.Append(ep->Block(), [&] {
auto* subgroup_invocation_id = b.Access(ty.u32(), str_param, 0_i);
auto* subgroup_size = b.Access(ty.u32(), str_param, 1_i);
auto* r = b.Multiply(ty.u32(), subgroup_invocation_id, subgroup_size);
b.Return(ep, r);
});
auto* src = R"(
Inputs = struct @align(4) {
id:u32 @offset(0), @builtin(subgroup_invocation_id)
size:u32 @offset(4), @builtin(subgroup_size)
}
%foo = @fragment func(%inputs:Inputs):u32 {
$B1: {
%3:u32 = access %inputs, 0i
%4:u32 = access %inputs, 1i
%5:u32 = mul %3, %4
ret %5
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Inputs = struct @align(4) {
id:u32 @offset(0)
size:u32 @offset(4)
}
foo_outputs = struct @align(4) {
tint_symbol:u32 @offset(0)
}
%foo_inner = func(%inputs:Inputs):u32 {
$B1: {
%3:u32 = access %inputs, 0i
%4:u32 = access %inputs, 1i
%5:u32 = mul %3, %4
ret %5
}
}
%foo = @fragment func():foo_outputs {
$B2: {
%7:u32 = hlsl.WaveGetLaneIndex
%8:u32 = hlsl.WaveGetLaneCount
%9:Inputs = construct %7, %8
%10:u32 = call %foo_inner, %9
%11:foo_outputs = construct %10
ret %11
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_NumWorkgroups_NonStruct) {
auto* num_workgroups = b.FunctionParam("num_wgs", ty.vec3<u32>());
num_workgroups->SetBuiltin(core::BuiltinValue::kNumWorkgroups);
auto* ep = b.Function("foo", ty.vec3<u32>(), core::ir::Function::PipelineStage::kCompute);
ep->SetParams({num_workgroups});
ep->SetWorkgroupSize(1, 1, 1);
b.Append(ep->Block(), [&] {
auto* r = b.Multiply(ty.vec3<u32>(), num_workgroups, num_workgroups);
b.Return(ep, r);
});
auto* src = R"(
%foo = @compute @workgroup_size(1, 1, 1) func(%num_wgs:vec3<u32> [@num_workgroups]):vec3<u32> {
$B1: {
%3:vec3<u32> = mul %num_wgs, %num_wgs
ret %3
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
foo_outputs = struct @align(16) {
tint_symbol:vec3<u32> @offset(0)
}
$B1: { # root
%tint_num_workgroups:ptr<uniform, vec3<u32>, read> = var @binding_point(0, 0)
}
%foo_inner = func(%num_wgs:vec3<u32>):vec3<u32> {
$B2: {
%4:vec3<u32> = mul %num_wgs, %num_wgs
ret %4
}
}
%foo = @compute @workgroup_size(1, 1, 1) func():foo_outputs {
$B3: {
%6:vec3<u32> = load %tint_num_workgroups
%7:vec3<u32> = call %foo_inner, %6
%8:foo_outputs = construct %7
ret %8
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_NumWorkgroups_Struct) {
auto* str_ty = ty.Struct(mod.symbols.New("Inputs"),
{
{
mod.symbols.New("num_wgs"),
ty.vec3<u32>(),
core::IOAttributes{
/* location */ std::nullopt,
/* blend_src */ std::nullopt,
/* color */ std::nullopt,
/* builtin */ core::BuiltinValue::kNumWorkgroups,
/* interpolation */ std::nullopt,
/* invariant */ false,
},
},
});
auto* str_param = b.FunctionParam("inputs", str_ty);
auto* ep = b.Function("foo", ty.vec3<u32>(), core::ir::Function::PipelineStage::kCompute);
ep->SetParams({str_param});
ep->SetWorkgroupSize(1, 1, 1);
b.Append(ep->Block(), [&] {
auto* num_workgroups = b.Access(ty.vec3<u32>(), str_param, 0_i);
auto* r = b.Multiply(ty.vec3<u32>(), num_workgroups, num_workgroups);
b.Return(ep, r);
});
auto* src = R"(
Inputs = struct @align(16) {
num_wgs:vec3<u32> @offset(0), @builtin(num_workgroups)
}
%foo = @compute @workgroup_size(1, 1, 1) func(%inputs:Inputs):vec3<u32> {
$B1: {
%3:vec3<u32> = access %inputs, 0i
%4:vec3<u32> = mul %3, %3
ret %4
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
Inputs = struct @align(16) {
num_wgs:vec3<u32> @offset(0)
}
foo_outputs = struct @align(16) {
tint_symbol:vec3<u32> @offset(0)
}
$B1: { # root
%tint_num_workgroups:ptr<uniform, vec3<u32>, read> = var @binding_point(0, 0)
}
%foo_inner = func(%inputs:Inputs):vec3<u32> {
$B2: {
%4:vec3<u32> = access %inputs, 0i
%5:vec3<u32> = mul %4, %4
ret %5
}
}
%foo = @compute @workgroup_size(1, 1, 1) func():foo_outputs {
$B3: {
%7:vec3<u32> = load %tint_num_workgroups
%8:Inputs = construct %7
%9:vec3<u32> = call %foo_inner, %8
%10:foo_outputs = construct %9
ret %10
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_NumWorkgroups_ExplicitBinding) {
auto* num_workgroups = b.FunctionParam("num_wgs", ty.vec3<u32>());
num_workgroups->SetBuiltin(core::BuiltinValue::kNumWorkgroups);
auto* ep = b.Function("foo", ty.vec3<u32>(), core::ir::Function::PipelineStage::kCompute);
ep->SetParams({num_workgroups});
ep->SetWorkgroupSize(1, 1, 1);
b.Append(ep->Block(), [&] {
auto* r = b.Multiply(ty.vec3<u32>(), num_workgroups, num_workgroups);
b.Return(ep, r);
});
auto* src = R"(
%foo = @compute @workgroup_size(1, 1, 1) func(%num_wgs:vec3<u32> [@num_workgroups]):vec3<u32> {
$B1: {
%3:vec3<u32> = mul %num_wgs, %num_wgs
ret %3
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
foo_outputs = struct @align(16) {
tint_symbol:vec3<u32> @offset(0)
}
$B1: { # root
%tint_num_workgroups:ptr<uniform, vec3<u32>, read> = var @binding_point(1, 23)
}
%foo_inner = func(%num_wgs:vec3<u32>):vec3<u32> {
$B2: {
%4:vec3<u32> = mul %num_wgs, %num_wgs
ret %4
}
}
%foo = @compute @workgroup_size(1, 1, 1) func():foo_outputs {
$B3: {
%6:vec3<u32> = load %tint_num_workgroups
%7:vec3<u32> = call %foo_inner, %6
%8:foo_outputs = construct %7
ret %8
}
}
)";
ShaderIOConfig config;
config.num_workgroups_binding = {1u, 23u};
Run(ShaderIO, config);
EXPECT_EQ(expect, str());
}
TEST_F(HlslWriterTransformTest, ShaderIOParameters_NumWorkgroups_AutoBinding) {
auto* num_workgroups = b.FunctionParam("num_wgs", ty.vec3<u32>());
num_workgroups->SetBuiltin(core::BuiltinValue::kNumWorkgroups);
auto* ep = b.Function("foo", ty.vec3<u32>(), core::ir::Function::PipelineStage::kCompute);
ep->SetParams({num_workgroups});
ep->SetWorkgroupSize(1, 1, 1);
b.Append(ep->Block(), [&] {
auto* r = b.Multiply(ty.vec3<u32>(), num_workgroups, num_workgroups);
b.Return(ep, r);
});
b.Append(mod.root_block, [&] {
for (uint32_t group = 0; group < 10; ++group) {
auto* v = b.Var<core::AddressSpace::kStorage, i32>();
v->SetBindingPoint(group, group + 1u);
}
});
auto* src = R"(
$B1: { # root
%1:ptr<storage, i32, read_write> = var @binding_point(0, 1)
%2:ptr<storage, i32, read_write> = var @binding_point(1, 2)
%3:ptr<storage, i32, read_write> = var @binding_point(2, 3)
%4:ptr<storage, i32, read_write> = var @binding_point(3, 4)
%5:ptr<storage, i32, read_write> = var @binding_point(4, 5)
%6:ptr<storage, i32, read_write> = var @binding_point(5, 6)
%7:ptr<storage, i32, read_write> = var @binding_point(6, 7)
%8:ptr<storage, i32, read_write> = var @binding_point(7, 8)
%9:ptr<storage, i32, read_write> = var @binding_point(8, 9)
%10:ptr<storage, i32, read_write> = var @binding_point(9, 10)
}
%foo = @compute @workgroup_size(1, 1, 1) func(%num_wgs:vec3<u32> [@num_workgroups]):vec3<u32> {
$B2: {
%13:vec3<u32> = mul %num_wgs, %num_wgs
ret %13
}
}
)";
EXPECT_EQ(src, str());
auto* expect = R"(
foo_outputs = struct @align(16) {
tint_symbol:vec3<u32> @offset(0)
}
$B1: { # root
%1:ptr<storage, i32, read_write> = var @binding_point(0, 1)
%2:ptr<storage, i32, read_write> = var @binding_point(1, 2)
%3:ptr<storage, i32, read_write> = var @binding_point(2, 3)
%4:ptr<storage, i32, read_write> = var @binding_point(3, 4)
%5:ptr<storage, i32, read_write> = var @binding_point(4, 5)
%6:ptr<storage, i32, read_write> = var @binding_point(5, 6)
%7:ptr<storage, i32, read_write> = var @binding_point(6, 7)
%8:ptr<storage, i32, read_write> = var @binding_point(7, 8)
%9:ptr<storage, i32, read_write> = var @binding_point(8, 9)
%10:ptr<storage, i32, read_write> = var @binding_point(9, 10)
%tint_num_workgroups:ptr<uniform, vec3<u32>, read> = var @binding_point(10, 0)
}
%foo_inner = func(%num_wgs:vec3<u32>):vec3<u32> {
$B2: {
%14:vec3<u32> = mul %num_wgs, %num_wgs
ret %14
}
}
%foo = @compute @workgroup_size(1, 1, 1) func():foo_outputs {
$B3: {
%16:vec3<u32> = load %tint_num_workgroups
%17:vec3<u32> = call %foo_inner, %16
%18:foo_outputs = construct %17
ret %18
}
}
)";
Run(ShaderIO, ShaderIOConfig{});
EXPECT_EQ(expect, str());
}
} // namespace
} // namespace tint::hlsl::writer::raise